sparc64 bulk build report

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

sparc64 bulk build report

Landry Breuil-5
bulk build on sparc64-1.ports.openbsd.org
started on  Sun Oct 29 13:36:33 MDT 2017
finished at Sat Nov 4 13:13:13 MDT 2017
lasted 06D16h36m
done with kern.version=OpenBSD 6.2-current (GENERIC) #323: Sat Oct 28 21:27:53 MDT 2017

built packages:6972
Oct 29:190
Oct 30:244
Oct 31:167
Nov 1:160
Nov 2:574
Nov 3:1439
Nov 4:4197



build failures: 32
http://build-failures.rhaalovely.net//sparc64/2017-10-29/archivers/woff2.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/comms/gammu.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/devel/arm-none-eabi/gcc-linaro.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/devel/astyle.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/devel/cbmc.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/devel/dlib.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/devel/harfbuzz.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/devel/keystone.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/devel/reposurgeon.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/devel/spidermonkey52.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/cataclysm-dda.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/liblcf.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/lugaru.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/openrct2.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/tome4,-data.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/geo/pdal.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/geo/pgpointcloud.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/graphics/dcmtk.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/graphics/glm.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/lang/apl.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/lang/verilator.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/mail/isync.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/net/avahi,bootstrap,no_gui.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/print/poppler.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/security/encfs.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/security/sslscan,openssl.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/security/yara/main.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/sysutils/facter.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/sysutils/upower.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/textproc/link-grammar,,-main.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/x11/gnome/totem-pl-parser.log
http://build-failures.rhaalovely.net//sparc64/2017-10-29/x11/worker.log

recurrent failures
 failures/devel/cbmc.log
 failures/devel/keystone.log
 failures/games/liblcf.log
 failures/games/lugaru.log
 failures/games/openrct2.log
 failures/geo/pdal.log
 failures/geo/pgpointcloud.log
 failures/graphics/dcmtk.log
 failures/lang/apl.log
 failures/mail/isync.log
 failures/security/encfs.log
 failures/security/sslscan,openssl.log
 failures/sysutils/facter.log
 failures/textproc/link-grammar,,-main.log
 failures/x11/worker.log
new failures
+++ ls-failures Sat Nov  4 13:13:43 2017
+failures/archivers/woff2.log
+failures/comms/gammu.log
+failures/devel/arm-none-eabi/gcc-linaro.log
+failures/devel/astyle.log
+failures/devel/dlib.log
+failures/devel/harfbuzz.log
+failures/devel/reposurgeon.log
+failures/devel/spidermonkey52.log
+failures/games/cataclysm-dda.log
+failures/graphics/glm.log
+failures/lang/verilator.log
+failures/net/avahi,bootstrap,no_gui.log
+failures/print/poppler.log
+failures/security/yara/main.log
+failures/sysutils/upower.log
+failures/x11/gnome/totem-pl-parser.log
resolved failures
--- ../old/sparc64/last//ls-failures Sun Oct 15 05:01:05 2017
-failures/audio/grip.log
-failures/converters/k2pdfopt.log
-failures/devel/arm-none-eabi/gcc-linaro,aarch64.log
-failures/emulators/fuse-utils.log
-failures/emulators/gr-lida.log
-failures/graphics/gdcm.log
-failures/graphics/pqiv.log
-failures/graphics/slop.log
-failures/mail/kopano/core,-mapi.log
-failures/math/gnumeric.log
-failures/misc/gtk3mm-documentation.log
-failures/net/farstream.log
-failures/net/libnids.log
-failures/net/owncloudclient.log
-failures/net/telegram-purple.log
-failures/print/texlive/base.log
-failures/telephony/asterisk,imap,-calendar.log
-failures/www/ruby-capybara-webkit,ruby24.log
-failures/www/webkitgtk4.log
-failures/x11/nitrogen.log
-failures/x11/qt5/qt3d,-examples.log
-failures/x11/qt5/qtgraphicaleffects.log
-failures/x11/qt5/qtquickcontrols.log
-failures/x11/qt5/qtquickcontrols2,,-main.log
-failures/x11/xfce4/xfce4-pulseaudio.log
Base libs:
crypto.42.0 c.91.0

X libs:

Reply | Threaded
Open this post in threaded view
|

Re: sparc64 bulk build report

trondd-2
On Sat, November 4, 2017 3:13 pm, [hidden email] wrote:

> bulk build on sparc64-1.ports.openbsd.org
> started on  Sun Oct 29 13:36:33 MDT 2017
> finished at Sat Nov 4 13:13:13 MDT 2017
> lasted 06D16h36m
> done with kern.version=OpenBSD 6.2-current (GENERIC) #323: Sat Oct 28
> 21:27:53 MDT 2017
>
> http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/cataclysm-dda.log
>
> new failures
> +failures/games/cataclysm-dda.log
>

Cataclysm won't build with base gcc because of missing -std=c++11 which is
what seems to be happening here.

I chose to configure the build for clang over gcc out of ports.

Is there a quick flag to disable this on non-clang archs?  Would anyone
think I should make this work on gcc archs?  Still wouldn't mean it will
compile and run and I can't test it.

Tim.

Reply | Threaded
Open this post in threaded view
|

Re: sparc64 bulk build report

Jeremie Courreges-Anglas-2

Hi Tim,

On Sat, Nov 04 2017, "trondd" <[hidden email]> wrote:

> On Sat, November 4, 2017 3:13 pm, [hidden email] wrote:
>> bulk build on sparc64-1.ports.openbsd.org
>> started on  Sun Oct 29 13:36:33 MDT 2017
>> finished at Sat Nov 4 13:13:13 MDT 2017
>> lasted 06D16h36m
>> done with kern.version=OpenBSD 6.2-current (GENERIC) #323: Sat Oct 28
>> 21:27:53 MDT 2017
>>
>> http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/cataclysm-dda.log
>>
>> new failures
>> +failures/games/cataclysm-dda.log
>>
>
> Cataclysm won't build with base gcc because of missing -std=c++11 which is
> what seems to be happening here.
>
> I chose to configure the build for clang over gcc out of ports.
>
> Is there a quick flag to disable this on non-clang archs?  Would anyone
> think I should make this work on gcc archs?  Still wouldn't mean it will
> compile and run and I can't test it.

I think that everybody should make an effort to support stuff on all
archs*.  The transition to clang makes this a bit worse than before, but
people forget that they can test both clang and base gcc on their fast
amd64 machine.

So nope, you can build it, and you can test it. ;)

* when it is doable, obviously; some ports are highly arch-dependent.
--
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524 E7EE

Reply | Threaded
Open this post in threaded view
|

Re: sparc64 bulk build report

Jeremie Courreges-Anglas-2
On Sun, Nov 05 2017, Jeremie Courreges-Anglas <[hidden email]> wrote:

> Hi Tim,
>
> On Sat, Nov 04 2017, "trondd" <[hidden email]> wrote:
>> On Sat, November 4, 2017 3:13 pm, [hidden email] wrote:
>>> bulk build on sparc64-1.ports.openbsd.org
>>> started on  Sun Oct 29 13:36:33 MDT 2017
>>> finished at Sat Nov 4 13:13:13 MDT 2017
>>> lasted 06D16h36m
>>> done with kern.version=OpenBSD 6.2-current (GENERIC) #323: Sat Oct 28
>>> 21:27:53 MDT 2017
>>>
>>> http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/cataclysm-dda.log
>>>
>>> new failures
>>> +failures/games/cataclysm-dda.log
>>>
>>
>> Cataclysm won't build with base gcc because of missing -std=c++11 which is
>> what seems to be happening here.
>>
>> I chose to configure the build for clang over gcc out of ports.
>>
>> Is there a quick flag to disable this on non-clang archs?  Would anyone
>> think I should make this work on gcc archs?  Still wouldn't mean it will
>> compile and run and I can't test it.
>
> I think that everybody should make an effort to support stuff on all
> archs*.  The transition to clang makes this a bit worse than before, but
> people forget that they can test both clang and base gcc on their fast
> amd64 machine.
>
> So nope, you can build it, and you can test it. ;)

Landry rightfully pointed out that I could have been more helpful.  To
test a build with base gcc, I just use this:

  make clean all repackage CC=gcc CXX=g++

Obviously with a reinstall if you want to test runtime.

> * when it is doable, obviously; some ports are highly arch-dependent.

--
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524 E7EE

Reply | Threaded
Open this post in threaded view
|

Re: sparc64 bulk build report

trondd-2
In reply to this post by Jeremie Courreges-Anglas-2
On Sat, November 4, 2017 7:54 pm, Jeremie Courreges-Anglas wrote:

>
> Hi Tim,
>
> On Sat, Nov 04 2017, "trondd" <[hidden email]> wrote:
>> On Sat, November 4, 2017 3:13 pm, [hidden email] wrote:
>>> bulk build on sparc64-1.ports.openbsd.org
>>> started on  Sun Oct 29 13:36:33 MDT 2017
>>> finished at Sat Nov 4 13:13:13 MDT 2017
>>> lasted 06D16h36m
>>> done with kern.version=OpenBSD 6.2-current (GENERIC) #323: Sat Oct 28
>>> 21:27:53 MDT 2017
>>>
>>> http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/cataclysm-dda.log
>>>
>>> new failures
>>> +failures/games/cataclysm-dda.log
>>>
>>
>> Cataclysm won't build with base gcc because of missing -std=c++11 which
>> is
>> what seems to be happening here.
>>
>> I chose to configure the build for clang over gcc out of ports.
>>
>> Is there a quick flag to disable this on non-clang archs?  Would anyone
>> think I should make this work on gcc archs?  Still wouldn't mean it will
>> compile and run and I can't test it.
>
> I think that everybody should make an effort to support stuff on all
> archs*.  The transition to clang makes this a bit worse than before, but
> people forget that they can test both clang and base gcc on their fast
> amd64 machine.
>
> So nope, you can build it, and you can test it. ;)

Yes, I know I can test the different compilers, I already have.  I was
refering to your astrisk.  :)

Is 'COMPILER= gcc' the right way to fix this situation?  It builds with
base clang and devel/gcc/4.9.

Thanks.
Tim.

>
> * when it is doable, obviously; some ports are highly arch-dependent.
> --
> jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524
> E7EE
>


Reply | Threaded
Open this post in threaded view
|

Re: sparc64 bulk build report

Jeremie Courreges-Anglas-2
On Sat, Nov 04 2017, "trondd" <[hidden email]> wrote:

> On Sat, November 4, 2017 7:54 pm, Jeremie Courreges-Anglas wrote:
>>
>> Hi Tim,
>>
>> On Sat, Nov 04 2017, "trondd" <[hidden email]> wrote:
>>> On Sat, November 4, 2017 3:13 pm, [hidden email] wrote:
>>>> bulk build on sparc64-1.ports.openbsd.org
>>>> started on  Sun Oct 29 13:36:33 MDT 2017
>>>> finished at Sat Nov 4 13:13:13 MDT 2017
>>>> lasted 06D16h36m
>>>> done with kern.version=OpenBSD 6.2-current (GENERIC) #323: Sat Oct 28
>>>> 21:27:53 MDT 2017
>>>>
>>>> http://build-failures.rhaalovely.net//sparc64/2017-10-29/games/cataclysm-dda.log
>>>>
>>>> new failures
>>>> +failures/games/cataclysm-dda.log
>>>>
>>>
>>> Cataclysm won't build with base gcc because of missing -std=c++11 which
>>> is
>>> what seems to be happening here.
>>>
>>> I chose to configure the build for clang over gcc out of ports.
>>>
>>> Is there a quick flag to disable this on non-clang archs?  Would anyone
>>> think I should make this work on gcc archs?  Still wouldn't mean it will
>>> compile and run and I can't test it.
>>
>> I think that everybody should make an effort to support stuff on all
>> archs*.  The transition to clang makes this a bit worse than before, but
>> people forget that they can test both clang and base gcc on their fast
>> amd64 machine.
>>
>> So nope, you can build it, and you can test it. ;)
>
> Yes, I know I can test the different compilers, I already have.  I was
> refering to your astrisk.  :)

Ah, ok. :)

> Is 'COMPILER= gcc' the right way to fix this situation?  It builds with
> base clang and devel/gcc/4.9.

The situation is muddy.  COMPILER = gcc is supported, but now we're not
using it for new ports.  You should pick stuff from base-clang (first in
the list), ports-gcc and ports-clang.

> Thanks.
> Tim.
>
>>
>> * when it is doable, obviously; some ports are highly arch-dependent.
>> --
>> jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524
>> E7EE
>>
>
>

--
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524 E7EE